Adaptive predistortion system
First Claim
1. A method for adaptively compensating a digital signal, comprising the following steps:
- accumulating statistical information at a receiver regarding a received signal transmitted by a transmitter, said accumulating step comprising the further steps of accumulating symbol cluster statistical information, accumulating weighted sums of the symbol coordinates to produce first, second and third matrices, inverting said first matrix and multiplying said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster, and multiplying said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster;
determining constellation distortions due to at least one voltage offset, gain mismatch ratio, lock angle error, quad angle error, nonlinear channel distortions and receiver distortions;
compensating for said constellation distortions in at least one of the in-phase and quadrature signal components of said signal;
processing said statistical information to determine compensation values; and
using said compensation values to compensate said signal.
2 Assignments
0 Petitions
Accused Products
Abstract
A method and apparatus are described for adaptively predistorting a signal before it is transmitted by a transmitter in order to compensate for nonlinearities which are introduced. Statistical information regarding the received signal is accumulated at a receiver, and is preferably transmitted back to the transmitter for processing. The processing includes a constellation analysis to determine the distortion characteristics of the signal, which may include bias, gain imbalance, lock and quad angle errors, and nonlinear distortions including AM-AM and AM-PM conversion effects, with all of these phenomena occurring anywhere in the transmitter-channel-receiverchain. From analysis of the statistics of the distortions in the received signal constellation, adjustments are made in the transmitter and/or receiver parameters. The positions of the signal symbols in the transmitted constellation are thereby correctively predistorted to compensate for and ameliorate the distortions in the received signal, thereby improving the fidelity of the said received signal.
-
Citations
54 Claims
-
1. A method for adaptively compensating a digital signal, comprising the following steps:
-
accumulating statistical information at a receiver regarding a received signal transmitted by a transmitter, said accumulating step comprising the further steps of accumulating symbol cluster statistical information, accumulating weighted sums of the symbol coordinates to produce first, second and third matrices, inverting said first matrix and multiplying said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster, and multiplying said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster;
determining constellation distortions due to at least one voltage offset, gain mismatch ratio, lock angle error, quad angle error, nonlinear channel distortions and receiver distortions;
compensating for said constellation distortions in at least one of the in-phase and quadrature signal components of said signal;
processing said statistical information to determine compensation values; and
using said compensation values to compensate said signal. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26)
transmitting from said receiver to a transmitter which transmitted said digital signals received at said receiver, statistical and control information indicative of the distortions in the received signals; and
processing said statistical and control information to determine at least one of type, magnitude and direction of compensating predistortions which may be applied at the transmitter to predistort said signal prior to its transmission, whereby the received signal at the receiver will have substantially no distortions.
-
-
5. The method of claim 1, wherein said compensating step is performed at one of the transmitter and receiver.
-
6. The method of claim 1, further comprising the following steps:
-
using a memory-based predistorter to compensate said signal;
determining any residual nonlinear distortion still present in said signal; and
if so,adaptively predistorting said signal to substantially eliminate said residual nonlinear distortion.
-
-
7. The method of claim 1, wherein control information indicative of said constellation distortions is exchanged between said transmitter and said receiver over a narrow bandwidth control channel.
-
8. The method of claim 7, wherein parameters in at least one of said receiver and transmitter are adjusted to correct distortions in said constellation arising from previously set parameter values.
-
9. The method of claim 1, wherein the step of determining constellation distortions due to lock angle error comprises the following steps:
-
determining at least one of the magnitude and direction of the lock angle error in the constellation; and
shifting the carrier phase in at least one of the transmitter and receiver to rotate the constellation in a direction and an amount which reduces the lock angle error.
-
-
10. The method of claim 1, wherein the step of determining constellation distortions due to quad angle error comprises the following steps:
-
determining at least one of the magnitude and direction of the quad angle error; and
adjusting at least one of the modulator quadrature phase shifter in the transmitter and the demodulator quadrature phase shifter in the receiver in a direction and an amount to reduce the quad angle error.
-
-
11. The method of claim 1, wherein the step of determining constellation distortions due to voltage offset comprises the following steps:
-
determining the magnitude of the voltage offset; and
adjusting the bias voltage in at least one of the transmitter and receiver in a direction and an amount to reduce the voltage offset.
-
-
12. The method of claim 1, wherein the step of determining constellation distortions due to gain mismatch ratio comprises the following steps:
-
determining the magnitude of the gain mismatch ratio; and
adjusting the amplifier gains in at least one of the transmitter and receiver in a direction and an amount to reduce the gain mismatch ratio.
-
-
13. The method of claim 1, wherein the step of determining constellation distortions due to nonlinear channel distortion comprises the following steps:
-
determining the magnitude of the nonlinear channel distortion; and
adjusting the position of each symbol within the constellation in at least one of the transmitter and receiver in a direction and an amount to reduce the nonlinear channel distortion.
-
-
14. The method of claim 1, wherein said first and second coordinates are the x and y cartesian coordinate dimensions of said symbol clusters, respectively.
-
15. The method of claim 1, wherein the step of accumulating weighted sums is performed in accordance with the following relationships to produce said first, second and third matrices, designated A, Bx and By:
-
wherein 1 is defined as (i−
(M+1)/2) and m is defined as (j−
(M+1/2)), M is the number of symbols in each coordinate of the constellation and i and j are integers ranging from 1 to M which indicate the location of each cluster.
-
-
16. The method of claim 15, wherein said first and second coordinate coefficient matrices are of the form:
-
17. The method of claim 16, wherein said gain mismatch ratio RG is defined as the following ratio Bxx/Byy.
-
18. The method of claim 16, wherein said second and third order gain mismatch ratios RG2 and RG3 are defined as the following ratios, respectively, Cxx/Cyy and Dxx/Dyy.
-
19. The method of claim 16, wherein said lock angle error is determined in accordance with the following relationship:
-
20. The method of claim 16, wherein said quad angle error is determined in accordance with the following relationship:
-
21. The method of claim 16, wherein the coefficient parameters are used to define the following quantities:
-
22. The method of claim 21, wherein lock angle error is determined in accordance with the following relationship:
-
23. The method of claim 21, wherein the gain mismatch ratio is determined in accordance with the following relationship:
-
24. The method of claims 1 or 16, wherein said method is performed in a repeater circuit.
-
25. The method of claim 1, wherein said processing step is performed at one of said transmitter and receiver.
-
26. The method of claim 1, wherein compensating step is performed at one of said transmitter and receiver.
-
27. A method for adaptively compensating a digital signal which is being transmitted by a transmitter and received by a receiver, comprising the following steps:
-
modulating a digital signal at the transmitter according to a predetermined modulation scheme;
adaptively compensating said modulated signal at said transmitter in accordance with compensation values determined from statistical information accumulated at the receiver regarding the received signal;
transmitting said compensated modulated signal to said receiver;
receiving said compensated modulated signal at said receiver;
accumulating statistical information at said receiver regarding said received signal, said accumulating step comprising the further steps of accumulating symbol cluster statistical information and using said symbol cluster statistical information to accumulate constellation statistical information, accumulating weighted sums of the symbol coordinates to produce first, second and third matrices;
inverting said first matrix and multiplying said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster;
multiplying said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster;
processing said statistical information to determine compensation values for constellation distortions due to at least one of voltage offset, gain mismatch ratio, lock angle error, quad angle error, nonlinear channel distortions and receiver distortions;
wherein said adaptive compensation step compensates for said constellation distortions in at least one of the in-phase and quadrature signal components of said signal. - View Dependent Claims (28, 29, 30, 31)
-
-
32. A method for adaptively compensating a digital signal which is being transmitted by a transmitter and received by a receiver, comprising the following steps:
-
modulating a digital signal at the transmitter according to a predetermined modulation scheme;
adaptively compensating said modulated signal at said receiver in accordance with compensation values determined from statistical information accumulated at the receiver regarding the received signal;
transmitting said modulated signal to said receiver;
receiving said modulated signal at said receiver;
accumulating statistical information at said receiver regarding said received signal, said accumulating step comprising the further steps of accumulating symbol cluster statistical information and using said symbol cluster statistical information to accumulate constellation statistical information, accumulating weighted sums of the symbol coordinates to produce first, second and third matrices;
inverting said first matrix and multiplying said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster;
multiplying said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster;
processing said statistical information to determine compensation values for constellation distortions due to at least one of voltage offset, gain mismatch ratio, lock angle error, quad angle error, nonlinear channel distortions and receiver distortions;
wherein said adaptive compensation step compensates for said constellation distortions in at least one of the in-phase and quadrature signal components of said signal. - View Dependent Claims (33, 34, 35, 36)
-
-
37. An adaptive compensation system for compensating a digital signal, comprising:
-
a statistical accumulation circuit which accumulates statistical information at a receiver regarding a received signal transmitted by a transmitter, said accumulation involving the accumulation of a symbol cluster statistical information which is used to accumulate constellation statistical information;
a compensation value determination circuit which processes said statistical information to determine compensation values; and
a compensation circuit which uses said compensation values to compensate said signal;
a constellation distortion estimation circuit which determines the constellation distortions due to at least one voltage offset, gain mismatch ratio, lock angle error, quad angle error, nonlinear channel distortions and receiver distortions;
wherein said compensation circuit compensates for said constellation distortions in at least one of the in-phase quadrature signal components of said signal; and
wherein said statistical accumulation circuit further comprises a weighted sum circuit which accumulates weighted sums of the symbol coordinates to produce first, second and third matrices, a matrix inversion circuit which inverts said first matrix and multiplies said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster, and a multiplication circuit which multiplies said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster. - View Dependent Claims (38, 39)
wherein 1 is defined as (i−
(M+1)/2) and m is defined as (j−
(M+1/2)), M is the number of symbols in each coordinate of the constellation and i and j are integers ranging from 1 to M which indicate the location of each cluster.
-
-
39. The system of claim 38, wherein said first and second coordinate coefficient matrices are of the form:
-
40. A system for adaptively compensating a digital signal which is being transmitted by a transmitter and received by a receiver, comprising:
-
a signal modulator which modulates a digital signal at the transmitter according to a predetermined modulation scheme;
an adaptive compensation circuit which compensates said modulated signal at said transmitter in accordance with compensation values determined from statistical information accumulated at the receiver regarding the received signal;
a transmitter for transmitting said compensated modulated signal to said receiver;
a receiving circuit which receives said compensated modulated signal at said receiver;
a statistical accumulation circuit which accumulates statistical information at said receiver regarding said received signal, said statistical accumulation circuit comprising a symbol cluster accumulation circuit which accumulates symbol cluster statistical information and uses said symbol cluster statistical information to accumulate constellation statistical information, a weighted sum circuit which accumulates weighted sums of the symbol coordinates to produce first, second and third matrices;
an inversion circuit which inverts said first matrix and multiplies said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster;
a multiplication circuit which multiplies said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster;
a processing circuit which processes said statistical information to determine compensation values for constellation distortions due to at least one of voltage offset, gain mismatch ratio, lock angle error, quad angle error, nonlinear channel distortions and receiver distortions;
wherein said adaptive compensation circuit compensates for said constellation distortions in at least one of the in-phase and quadrature signal components of said signal. - View Dependent Claims (41, 42, 43, 44)
-
-
45. A system for adaptively compensating a digital signal which is being transmitted by a transmitter and received by a receiver, comprising:
-
a signal modulator which modulates a digital signal at the transmitter according to a predetermined modulation scheme;
an adaptive compensation circuit which adaptively compensates said modulated signal at said receiver in accordance with compensation values determined from statistical information accumulated at the receiver regarding the received signal;
a transmitter for transmitting said modulated signal to said receiver;
a receiving circuit which receives said modulated signal at said receiver;
a statistical accumulation circuit which accumulates statistical information at said receiver regarding said received signal, said statistical accumulation circuit comprising;
a symbol cluster accumulation circuit which accumulates symbol cluster statistical information and uses said symbol cluster statistical information to accumulate constellation statistical information, a weighted sum circuit which accumulates weighted sums of the symbol coordinates to produce first, second and third matrices;
an inversion circuit which inverts said first matrix and multiplying said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster;
a multiplication circuit which multiplies said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster;
a processing circuit which processes said statistical information to determine compensation values for constellation distortions due to at least one of voltage offset, gain mismatch ratio, lock angle error, quad angle error, nonlinear channel distortions and receiver distortions;
wherein said adaptive compensation circuit compensates for said constellation distortions in at least one of the in-phase and quadrature signal components of said signal. - View Dependent Claims (46, 47, 48, 49)
-
-
50. A method for adaptively compensating a digital signal, comprising the following steps:
-
accumulating statistical information at a receiver regarding a received signal transmitted by a transmitter, said accumulating step comprising the further steps of accumulating symbol cluster statistical information, using said symbol cluster statistical information to accumulate constellation statistical information, accumulating weighted sums of the symbol coordinates to produce first, second and third matrices, inverting said first matrix and multiplying said inverted first matrix by said second matrix to produce a first coefficient matrix of the coefficients for a first coordinate of the constellation used in a least squares model for estimating the center of each symbol cluster, multiplying said inverted first matrix by said third matrix to produce a second coefficient matrix of the coefficients for a second coordinate of the constellation used in said least squares model for estimating the center of each symbol cluster;
processing said statistical information to determine compensation values; and
using said compensation values to compensate said signal. - View Dependent Claims (51, 52, 53, 54)
transmitting from said receiver to a transmitter which transmitted said digital signals received at said receiver, statistical and control information indicative of the distortions in the received signals; and
processing said statistical and control information to determine at least one type, magnitude and direction of compensating predistortions which may be applied at the transmitter to predistort said signal prior to its transmission, whereby the received signal at the receiver will have substantially no distortions.
-
-
54. The method of claim 50, further comprising the following steps:
-
using a memory-based predistorter to compensate said signal;
determining any residual nonlinear distortion still present in said signal; and
if so,adaptively predistorting said signal to substantially eliminate said residual nonlinear distortion.
-
Specification